There is still a great deal that is unclear about the ways that physiological patterns of estrogen secretion from the ovary influence the female brain, under normal conditions and in disease. In the last project period, we made several strides forward, focusing on the hippocampus because of the substantial evidence that estrogen exerts robust effects on hippocampal neurons. First, we showed that the normal preovulatory surge in estrogen in the adult female rat is associated with robust increases in hippocampal excitability, especially in area CA3. These studies were the first to show a physiological correlate in hippocampus to gonadal estrogen secretion, and identified CA3 as a potential target. We also suggested that the mechanism of estrogen action was based on one of its target genes, the neurotrophin BDNF (brain-derived neurotrophic factor), which is concentrated in a major glutamatergic input to the CA3 region known as the mossy fiber pathway. We then devised the first estrogen replacement paradigm in the ovariectomized rat to accurately simulate the serum changes in estradiol during the preovulatory estrogen surge, and confirmed that estrogen was responsible for the effects we had documented in the intact rat. We began to elucidate the in vivo implications for memory based on our novel in vivo finding that estrogen increased hippocampal sharp waves, a synchronous discharge of CA3 neurons that is known to be important to memory consolidation. In addition, we developed preliminary data to support a role of another target gene of estrogen in the regulation of hippocampal excitability: vascular endothelial growth factor (VEGF). We now propose that the secretion of estradiol from the ovary leads to upregulation of BDNF and VEGF. BDNF increases excitablity primarily in pyramidal cells, and VEGF controls BDNF-mediated excitability by inhibiting synaptic transmission. Together, estrogen leads to improved hippocampal function by increasing BDNF, and VEGF keeps excitability in check, preventing seizures. One of the aspects of our preliminary data that is intriguing is the relatively long-lasting elevation in BDNF compared to estradiol and VEGF. This observation may be relevant to women with epilepsy, because a relatively high BDNF/VEGF ratio may explain seizures that increase after estrogen rises during the menstrual cycle in women with epilepsy. Using a rat model of epilepsy in females that preserves cyclical ovarian secetion, a model we have developed in the past period of funding, we will test the hypothesis. If we are correct, our results will provide a new framework for understanding how estrogen influences hippocampus, and catamenial epilepsy. Moreover, our data will provide support for new anticonvulsant strategies for women with epilepsy based on inhibiting BDNF or enhancing VEGF.